Can you please confirm that psi sdm must not be used to analyse one sided studies because the MetaNSUE algorithm considered only censorship by statistical significance, and not by effect sign. In other words, you cannot analyse activations only. 

Otherwise, could you please expand on how the analysis considered censorship of effect sign? 

Thanks

Dear Chris,

SDM-PSI considers both positive and negative effects, and it gives you two maps for each sign. You cannot censor the effect of the signs. You can only focus on the results of the sign you want (the best approach) or only use as input studies with that sign (could be biased)

Best,

Lydia

Sorry, I think there is a misunderstanding. I'm talking about the studies being analysed. They must provide both effect signs, or they cannot be included in the analysis. Therefore, it's impossible to meta analyse studies reporting positive effects only, as is commonly donewith other methods?